Fault tolerant unsupervised kernel-based information clustering in hyperspectral images
- 1 July 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2191-2194
Abstract
In this work we derive a novel clustering scheme for hyperspectral pixels according to the material they sense. We utilize statistical correlations that pixels sensing the same material exhibit. Specifically, kernel learning is combined with a norm-one regularized canonical correlations framework that can perform data clustering on nonlinearly dependent data. To tackle the derived minimization formulation we employ gradient descent iterations that enable a computationally efficient determination of proper sparse clustering matrices. Extensive numerical tests on real hyperspectral images reveal that the proposed approach, in spite of being unsupervised, can outperform existing supervised and unsupervised techniques especially in the presence of missing pixels that may be caused by malfunctioning in the data acquisition system.Keywords
This publication has 7 references indexed in Scilit:
- Online Distributed Sparsity-Aware Canonical Correlation AnalysisIEEE Transactions on Signal Processing, 2015
- Spectral Nonlocal Restoration of Hyperspectral Images With Low-Rank PropertyIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014
- Generalized Composite Kernel Framework for Hyperspectral Image ClassificationIEEE Transactions on Geoscience and Remote Sensing, 2013
- Kernel-based methods for hyperspectral image classificationIEEE Transactions on Geoscience and Remote Sensing, 2005
- Robust support vector method for hyperspectral data classification and knowledge discoveryIEEE Transactions on Geoscience and Remote Sensing, 2004
- Combination of neural and statistical algorithms for supervised classification of remote-sensing imagesPattern Recognition Letters, 2000
- Least squares quantization in PCMIEEE Transactions on Information Theory, 1982